Hugging Face's logo

Doctor-Shotgun
/
Norobara-ZLoss-8x7b-qlora

PEFT
Safetensors
mixtral
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files
Norobara-ZLoss-8x7b-qlora / README.md
Doctor-Shotgun's picture
Doctor-Shotgun
Upload folder using huggingface_hub
b8d1913
preview code
|
raw
history blame
3.96 kB
metadata
library_name: peft
tags:
  - generated_from_trainer
base_model: mistralai/Mixtral-8x7B-v0.1
model-index:
  - name: mixtral-norobara-qlora-out
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.3.0

base_model: mistralai/Mixtral-8x7B-v0.1
model_type: AutoModelForCausalLM
tokenizer_type: LlamaTokenizer
trust_remote_code: true

model_config:
  output_router_logits: true
  router_aux_loss_coef: 0.02
  router_z_loss_coef: 0.001

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: capybara-sharegpt.jsonl
    type: sharegpt
    conversation: alpaca_multiturn
  - path: no-robots-sharegpt-fixed.jsonl
    type: sharegpt
    conversation: alpaca_multiturn
  - path: toxicsharegpt-NoWarning.jsonl
    type: sharegpt
    conversation: alpaca_multiturn
  - path: camel-verified-sharegpt.jsonl
    type: sharegpt
    conversation: alpaca_multiturn
dataset_prepared_path: last_run_prepared
val_set_size: 0.0
output_dir: mixtral-norobara-qlora-out

## You can optionally freeze the entire model and unfreeze a subset of parameters
#unfrozen_parameters:
#  - lm_head.*
#  - model.embed_tokens.*
#  - model.layers.2[0-9]+.block_sparse_moe.gate.*
#  - model.layers.2[0-9]+.block_sparse_moe.experts.*
#  - model.layers.3[0-9]+.block_sparse_moe.gate.*
#  - model.layers.3[0-9]+.block_sparse_moe.experts.*

adapter: qlora
lora_model_dir:

sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true

lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:
#lora_target_modules:
#  - gate
#  - q_proj
#  - k_proj
#  - v_proj
#  - o_proj
#  - w1
#  - w2
#  - w3

wandb_project: mixtral-lora
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 16
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0001

train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: true

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

loss_watchdog_threshold:
loss_watchdog_patience:

warmup_steps: 10
evals_per_epoch:
eval_table_size:
eval_table_max_new_tokens:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

mixtral-norobara-qlora-out

This model was trained from scratch on the None dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 16
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 10
  • num_epochs: 3

Training results

Framework versions

  • Transformers 4.37.0.dev0
  • Pytorch 2.1.2+cu121
  • Datasets 2.16.1
  • Tokenizers 0.15.0

Training procedure

The following bitsandbytes quantization config was used during training:

  • quant_method: bitsandbytes
  • load_in_8bit: False
  • load_in_4bit: True
  • llm_int8_threshold: 6.0
  • llm_int8_skip_modules: None
  • llm_int8_enable_fp32_cpu_offload: False
  • llm_int8_has_fp16_weight: False
  • bnb_4bit_quant_type: nf4
  • bnb_4bit_use_double_quant: True
  • bnb_4bit_compute_dtype: bfloat16

Framework versions

  • PEFT 0.6.0